Adam Gleave
Adam Gleave
Thanks for putting the PDF together Lev! I don't follow the 2nd step in lemma 2, where the denominator switches from 1-alpha to 1-w_{m,m}. I thought w_{m,m} was alpha/b --...
Ah, I guess this error gets "canceled out" later on: the weights sum to 1/b not 1, so if we just change the 2nd step to be 1 - b*w_{m,m}...
This is fairly standard practice for Gym, e.g. this happens in [Gym itself](https://github.com/openai/gym/blob/master/gym/envs/__init__.py#L11) and [roboschool](https://github.com/openai/roboschool/blob/master/roboschool/__init__.py). Of course, standard practice is not necessarily the same as *good* practice. The main reasons...
Sorry this seems to have fallen by the wayside. I'm happy to review this if it's brought up to date with current `master`. Do we still need the SB3 fork...
Thanks for the summary Sam! > I'm in the process of (belatedly) bringing our SB3 and imitation branches even with master for the IL representations project. Here is a diff...
Closing in favor of #519
One useful tool might be [airspeed velocity (asv)](https://pypi.org/project/asv/) to keep track of metrics over time.
@taufeeque9 has been working on this, but it's a big enough task it might make sense to split it up (e.g. you each take some subset of algorithms and work...
Thanks for implementing this! Feel free to request my review when finished, but I don't want to block this and will be largely unavailable Mon-Wed, so happy to defer to...
Thanks, will aim to review tomorrow (Friday) or Monday.